View PDF Version

No. 03-9685

In the Supreme Court of the United States

ROBERT JOHNSON, JR., PETITIONER

v.

UNITED STATES OF AMERICA

ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT

BRIEF FOR THE UNITED STATES

PAUL D. CLEMENT
Acting Solicitor General
Counsel of Record
CHRISTOPHER A. WRAY
Assistant Attorney General
MICHAEL R. DREEBEN
Deputy Solicitor General
DAN HIMMELFARB
Assistant to the Solicitor
General
NINA GOODMAN
Attorney
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

QUESTION PRESENTED

Whether, when a federal prisoner whose sentence was enhanced on the basis of a state-court conviction files a motion to vacate the sentence on the ground that the state-court conviction has been vacated, the one-year limitation period under 28 U.S.C. 2255 para. 6(4) runs from the date on which the state-court conviction was vacated.

In the Supreme Court of the United States

No. 03-9685

ROBERT JOHNSON, JR., PETITIONER

v.

UNITED STATES OF AMERICA

ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT

BRIEF FOR THE UNITED STATES

OPINIONS BELOW

The opinion of the court of appeals (J.A. 22-46) is reported at 340 F.3d 1219. The order of the district court (J.A. 18-20) is unreported.

JURISDICTION

The judgment of the court of appeals was entered on August 5, 2003. A petition for rehearing was denied on December 22, 2003 (J.A. 48-53). The petition for a writ of certiorari was filed on March 22, 2004, and granted on September 28, 2004. The jurisdiction of this Court rests on 28 U.S.C. 1254(1).

STATUTORY PROVISION INVOLVED

Paragraph 6 of Section 2255 of Title 28 of the United States Code provides as follows:

A 1-year period of limitation shall apply to a motion under this section. The limitation period shall run from the latest of-

(1) the date on which the judgment of conviction becomes final;

(2) the date on which the impediment to making a motion created by governmental action in violation of the Constitution or laws of the United States is removed, if the movant was prevented from making a motion by such governmental action;

(3) the date on which the right asserted was initially recognized by the Supreme Court, if that right has been newly recognized by the Supreme Court and made retroactively applicable to cases on collateral review; or

(4) the date on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence.

STATEMENT

1. a. Various provisions of federal law require a district court to impose an enhanced sentence if the defendant has prior convictions for certain types of offenses. For example, under the Armed Career Criminal Act of 1984 (ACCA), 18 U.S.C. 924(e), defendants found guilty of certain firearms offenses are subject to a mandatory minimum prison term of 15 years if they have three prior convictions for a "violent felony" or "serious drug offense." Similarly, under the "career offender" provision of the United States Sentencing Guidelines, adult defendants found guilty of a felony "crime of violence" or "controlled substance offense" are subject to an enhanced offense level and automatic placement in criminal history category VI if they have at least two prior felony convictions for a crime of violence or controlled substance offense. Sentencing Guidelines § 4B1.1.

Defendants sentenced under laws of this type frequently file collateral attacks on the validity of a prior state conviction used to enhance the federal sentence. This Court has twice addressed the question of where a defendant sentenced under the ACCA may bring such a challenge. In Custis v. United States, 511 U.S. 485 (1994), the Court held that the federal sentencing proceeding is not the proper forum for a collateral attack on a prior state conviction. In Daniels v. United States, 532 U.S. 374 (2001), the Court held that a post-conviction proceeding under 28 U.S.C. 2255 is not the proper forum.1 Since this Court's decision in Custis, the courts of appeals have uniformly concluded that a federal prisoner who has successfully attacked a prior state conviction in an appropriate forum (generally a post-conviction proceeding in state court) may bring a Section 2255 motion challenging a federal sentence that was enhanced on the basis of the since-vacated state conviction. See, e.g., United States v. Doe, 239 F.3d 473, 475 (2d Cir. 2001) (per curiam) (citing cases).

Federal prisoners who seek to have a prior state-court conviction vacated in order to challenge a federal recidivist sentence often file the state-court motion many years after the federal sentence has become final. Delaying a challenge to the state conviction can dramatically increase the likelihood of success, since a state prosecutor may be either unable to defend an old conviction (because of the difficulty or impossibility of locating transcripts and other relevant records) or unwilling to do so (because of the likelihood that the defendant will have long since completed his state sentence and that vacating the conviction will immediately affect only the length of his federal sentence). In a case that typifies this situation, a federal prisoner's "state court collateral attack upon a nineteen-year old state manslaughter sentence was not contested by the state executives at hearing." United States v. Walker, 198 F.3d 811, 814 (11th Cir. 1999) (Hill, J., concurring).

b. Before the enactment of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), Pub. L. No. 104-132, 110 Stat. 1214, no statute of limitations governed motions for collateral relief under 28 U.S.C. 2255. Under the pre-AEDPA law, a federal prisoner could file such a motion "at any time," 28 U.S.C. 2255 para. 2 (1994), and the government was entitled to have the motion dismissed as untimely only if it was "prejudiced in its ability to respond to the motion by delay in its filing," Rule 9(a) of the Rules Governing § 2255 Proceedings. As a result, a prisoner could "wait a decade" or more after his conviction before seeking collateral relief. Lindh v. Murphy, 96 F.3d 856, 866 (7th Cir. 1996) (en banc), rev'd on other grounds, 521 U.S. 320 (1997).

To "curb the lengthy delays in filing" that "often occur in federal habeas corpus litigation," H.R. Rep. No. 23, 104th Cong., 1st Sess. 9 (1995), Congress added a statute of limitations when it enacted the AEDPA. Section 105 of the law, 110 Stat. 1220, which is codified in paragraph 6 of 28 U.S.C. 2255, establishes a "1-year period of limitation" for motions brought under Section 2255. The period runs from "the latest" of a number of events, which are set forth in subparagraphs (1) through (4) of paragraph 6.

The general rule, found in subparagraph (1), and interpreted by this Court in Clay v. United States, 537 U.S. 522 (2003), is that the triggering date is "the date on which the judgment of conviction becomes final."2 Three exceptions to that rule permit the one-year limitation period to begin later. Under the exception found in subparagraph (2), the one-year period begins to run on "the date on which the impediment to making a motion created by governmental action in violation of the Constitution or laws of the United States is removed, if the movant was prevented from making a motion by such governmental action." Under the exception found in subparagraph (3), the provision at issue in Dodd v. United States, cert. granted, No. 04-5286 (Nov. 29, 2004), the one-year period begins to run on "the date on which the right asserted was initially recognized by the Supreme Court, if that right has been newly recognized by the Supreme Court and made retroactively applicable to cases on collateral review." This case involves the exception found in subparagraph (4), under which the one-year period begins to run on "the date on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence."3

2. On six occasions in 1993 and 1994, petitioner sold cocaine base to undercover law-enforcement officers in Lenox, Georgia. Petitioner made some of the sales while he was on weekend furloughs from the Cook County, Georgia, Jail, where he was serving a sentence for a prior state drug offense. PSR ¶¶ 9-15, 60-61.

3. A grand jury in the Middle District of Georgia returned an indictment charging petitioner with one count of conspiracy to possess cocaine base with the intent to distribute it, in violation of 21 U.S.C. 846, and five counts of distribution of cocaine base, in violation of 21 U.S.C. 841(a)(1) and 18 U.S.C. 2. Pursuant to a plea agreement, petitioner pleaded guilty to a single count of distribution. Gov't C.A. Br. 2; PSR ¶¶ 1-2, 5-6.

In the Presentence Investigation Report (PSR), the Probation Office determined that petitioner had at least eight prior convictions, which resulted in a total of 20 criminal history points and placed him in criminal history category VI. PSR ¶¶ 39- 64. The Probation Office also determined that petitioner qualified as a career offender under Sentencing Guidelines § 4B1.1, based on two 1989 convictions in Cook County, Georgia, Superior Court for distribution of cocaine. PSR ¶¶ 32-33, 52-55, 65. In both cases, the PSR noted, petitioner had "waived counsel," pleaded guilty to the charge, and received a sentence of probation. PSR ¶¶ 52-55. Petitioner's offense level under the career-offender provision was 32 and, with a 3-level reduction for acceptance of responsibility, the applicable Guidelines range was 151 to 188 months of imprisonment. PSR ¶¶ 32-36, 79.

Petitioner filed written objections to the PSR, including an objection to his classification as a career offender, but he did not explain the basis for the objection. At the sentencing hearing, petitioner's counsel advised the district court that the objections to the PSR were being withdrawn, and petitioner confirmed that he did not wish to pursue his objections. The district court sentenced petitioner as a career offender, and imposed a prison term of 188 months. J.A. 23; Gov't C.A. Br. 4; PSR Add. 1.

4. On December 22, 1995, the court of appeals affirmed petitioner's sentence. J.A. 7-8. In his appeal, petitioner argued that his prior state-court convictions were based on guilty pleas entered without a valid waiver of his right to counsel, and that he therefore should not have been sentenced as a career offender. Ibid. Because petitioner had waived his objection to the career-offender classification at sentencing, the court of appeals declined to consider whether his prior convictions were valid. J.A. 8. On April 22, 1996, this Court denied certiorari. 517 U.S. 1162.

5. On February 6, 1998, petitioner filed a petition for a writ of habeas corpus in Wayne County, Georgia, Superior Court, challenging seven of his prior state convictions on the ground that he had not validly waived his right to counsel at the guilty plea hearings in those cases. J.A. 9, 24. On October 24, 2000, in a brief order, the trial court granted the petition and vacated petitioner's convictions. J.A. 9-10, 24. Noting that the State had responded to the petition by denying the allegations but had not "filed any further transcripts of the hearings," the Superior Court found that "the record in these cases does not show an affirmative waiver of [petitioner's] right to an attorney." J.A. 10. One of the seven convictions vacated by the state court was one of the two that served as the basis for petitioner's classification as a career offender. Compare J.A. 9 with PSR ¶¶ 33, 52, 54.

6. On February 13, 2001, petitioner filed a motion under 28 U.S.C. 2255, claiming that, because his state convictions had been vacated, he no longer qualified as a career offender. J.A. 24. Petitioner contended that his motion was timely because the state court's order granting his habeas corpus petition "constitute[d] new evidence which did not exist and could not have been discovered" earlier, and the one-year time limit therefore began to run on the date the court issued its ruling. Mem. in Support of Mot. to Vacate, Set Aside or Correct Sentence 3.

A magistrate judge recommended that petitioner's Section 2255 motion be denied as untimely. J.A. 11-17. The district court adopted the magistrate judge's report and recommendation and denied the motion, holding that the AEDPA's one-year limitation period expired on April 24, 1997, a year after petitioner's federal conviction became final by virtue of this Court's denial of certiorari. J.A. 18-20. The court noted that petitioner had waited 21 months from the denial of certiorari to file his state-court habeas corpus petition, and that "a period of almost five * * * years [had] passed before he eventually filed his Section 2255 motion." J.A. 19-20. The district court denied petitioner's motion for a certificate of appealability, but the court of appeals issued one, limited to the question whether petitioner's Section 2255 motion was timely. J.A. 4, 24.

7. A divided panel of the court of appeals affirmed. J.A. 21-46.

a. The majority rejected petitioner's claim that the one-year limitation period did not commence until October 24, 2000, when the state court vacated his prior convictions. Relying on the First Circuit's decision in Brackett v. United States, 270 F.3d 60 (2001), cert. denied, 535 U.S. 1003 (2002), see J.A. 27-31, the majority held that "a state court's vacatur of a federal prisoner's prior state convictions is not a 'fact supporting the claim or claims' under § 2255 ¶ 6(4) from which [the] AEDPA's statute of limitation will run," J.A. 33. Unlike the fact that petitioner did not validly "waive[] his right to counsel before pleading guilty to the state charges," J.A. 27, the majority explained, "the vacatur of prior state convictions is a court action obtained at the behest of a federal prisoner, not [a fact] 'discovered' by him," J.A. 28. Noting the AEDPA's "clear legislative purpose" of "ensur[ing] a greater degree of finality for convictions," J.A. 29, the majority also reasoned that petitioner's interpretation of Section 2255 para. 6(4) would "run directly counter to the general congressional intent behind [the] AEDPA and indefinitely extend the opportunity for post-conviction challenges," J.A. 30. Because petitioner "knew all of the facts supporting his challenge to his state convictions before his federal conviction became final," and because he filed his Section 2255 motion more than three years after the expiration of the one-year grace period running from the AEDPA's effective date, the majority determined that petitioner's motion was untimely. J.A. 33.

The majority also held that petitioner was not entitled to equitable tolling of the one-year limitation period, because he could not show that his delay in filing the Section 2255 motion was "the result of extraordinary circumstances that were beyond his control even with the exercise of due diligence." J.A. 34. The majority observed that a prisoner in a case of this type could be eligible for equitable tolling if he diligently pursued his state claim and was unable to obtain a decision from the state court before the expiration of the limitation period solely because of the court's delay. J.A. 37. In holding that equitable tolling was unavailable in this case, the court explained that petitioner "was plainly aware of whatever facts supported his collateral attack on his prior state convictions at the time his federal sentence became final," J.A. 34, and yet he "did nothing [before or] during the one-year AEDPA grace period to attack [the] prior convictions," J.A. 37.

b. Judge Roney dissented. J.A. 40-46. In his view, the limitation period under paragraph 6(4) runs from the date of the state court's order, because a decision vacating a prisoner's prior conviction is a "fact" that is not "discoverable" until the court issues its decision. J.A. 41. Judge Roney would have followed the Fourth Circuit's decision in United States v. Gadsen, 332 F.3d 224 (2003), so holding. J.A. 40.4

SUMMARY OF ARGUMENT

The court of appeals correctly held that petitioner's Section 2255 motion was untimely.

A. Petitioner contends that, when a federal prisoner challenges a recidivist sentence enhancement on the ground that the prior state conviction has been vacated, the one-year limitation period commences under 28 U.S.C. 2255 para. 6(4) on the date the state conviction is vacated. As an initial matter, petitioner's interpretation is inconsistent with the statutory text. Paragraph 6(4) provides that the limitation period runs from the date on which the "fact[] supporting the claim" in the Section 2255 motion "could have been discovered." A court order is obtained at the moving party's behest, not "discovered" by him.

Petitioner defends his interpretation on the ground that a court order can be "discovered" after it is issued. But that is not the type of discovery that Congress had in mind. Since a party will almost always know about a court order it has procured as soon as the order is issued, it would make little sense for a limitation period to be triggered by the discovery of the order. That view is confirmed by the fact that, under two other subparagraphs of Section 2255's limitation provision, the one-year period commences with the issuance, not the discovery, of a judicial decision. See 28 U.S.C. 2255 para. 6(1) (this Court's affirmance of conviction or denial of certiorari); id. para. 6(3) (this Court's initial recognition of right asserted). The triggering date under paragraph 6(4), moreover, is not the date on which the fact supporting the claim was discovered, but the date on which it could have been discovered "through the exercise of due diligence," a textual limitation that reflects Congress's intent that Section 2255 motions be filed expeditiously. Petitioner's interpretation, which requires a prisoner to exercise due diligence in discovering the order once it is issued, but not in taking the steps necessary to obtain the order, deprives the "due diligence" requirement of nearly all its force.

Petitioner's interpretation of paragraph 6(4) also undermines the purpose of Section 2255's limitation provision, which is to "reduce[] the potential for delay on the road to finality." Duncan v. Walker, 533 U.S. 167, 179 (2001). Under petitioner's interpretation, federal prisoners in cases of this type will have an incentive to increase the delay in challenging a prior state conviction, and thus in challenging the federal sentence, because the passage of time increases the likelihood that records will be unavailable and that the State will be uninterested in defending a conviction for which the sentence has long since been served.

Petitioner claims that his interpretation furthers the principles of comity and federalism, by ensuring that state courts are able to "exculpate criminal defendants." Br. 35. But petitioner will remain exculpated of the state crime whether or not his challenge to the federal sentence goes forward, and the question whether the federal sentence may be challenged is a matter of federal, not state, concern. Moreover, in a context in which the main effect of vacating a state conviction is on a federal sentence (so much so that state officials often have little incentive to oppose the state-court motion), principles of comity and federalism should not lead federal courts to ignore society's interest in the finality of federal sentences.

Under the correct interpretation of paragraph 6(4), the one-year limitation period runs from the date the facts that form the basis for the motion to vacate the prior state conviction were discoverable in the exercise of due diligence. Unlike the order granting the motion to vacate the state conviction, the facts that form the basis for the motion are both facts that can be "discovered" and facts that, if pursued with "due diligence," will ordinarily lead to the prompt filing of a Section 2255 motion.

Contrary to petitioner's contention, the government's interpretation of paragraph 6(4) does not make it "virtually impossible" (Br. 32) for a prisoner to comply with the statute of limitations. Petitioner suggests that the delay between the filing of a motion to vacate a state conviction and the order granting it will generally be more than one year. But the factual basis for the state motion is typically discoverable at the time of the asserted violation, which is usually years before the federal conviction becomes final, and even when it is not discoverable until the federal sentencing, the prisoner will still have the length of the federal direct-appeal process plus an additional year to challenge the state conviction. In a case in which the factual basis for the state claim is not discoverable until after the federal conviction has become final, the prisoner may still be able to obtain vacatur of the state conviction within a year. And in a case in which the state court's delay prevents a prisoner from obtaining vacatur of the state conviction within a year of the date the federal conviction became final or the date the factual basis for the state claim was discoverable, the prisoner may be eligible for equitable tolling.

B. Even if, as petitioner contends, the "fact[] supporting the claim" whose imputed discovery triggers the limitation period is the vacatur of the state conviction, the limitation period under paragraph 6(4) does not run from the date the vacatur was obtained. It runs from the date the vacatur could have been obtained if the prisoner had exercised due diligence. If the factual basis for the state claim could have been discovered earlier, or the state-court motion could have been filed earlier, then the date the court order vacating the state conviction could have been obtained, and thus "discovered," will be earlier than the date it was issued. This alternative interpretation is consistent with the text and purpose of paragraph 6(4), and it avoids petitioner's principal objections to the other interpretation.

C. Under either interpretation of paragraph 6(4), petitioner's Section 2255 motion was untimely. If the limitation period under that provision runs from the date the factual basis for the state-court motion could have been discovered, petitioner's Section 2255 motion was filed nearly four years too late, because he was aware of the factual basis for the state claim before the AEDPA's effective date (the triggering date under paragraph 6(1)) and the Section 2255 motion was filed nearly five years afterwards. If the limitation period under paragraph 6(4) runs from the date the vacatur of the state conviction could have been discovered with due diligence, petitioner's Section 2255 motion was untimely because the factual basis for petitioner's state claim was discoverable by December 1995 (at the latest); petitioner could have filed his state-court motion by December 1996 (and probably would not have required a full year to do so); the motion could have been granted, and thus "discovered," by August 1999 (assuming the court took the same amount of time to decide the motion that it took when the motion was filed); and petitioner did not file his Section 2255 motion until February 2001.

ARGUMENT

PETITIONER'S MOTION UNDER 28 U.S.C. 2255 WAS UNTIMELY

A motion under 28 U.S.C. 2255 must be filed within one year of the latest of four dates, including "the date on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence." 28 U.S.C. 2255 para. 6(4). Petitioner's Section 2255 motion challenges his federal sentence on the ground that a prior state conviction used to enhance the sentence has been vacated. He contends that the one-year limitation period under paragraph 6(4) of Section 2255 ran from the date the state court vacated the prior conviction.

Petitioner is mistaken. In a case of this type, the "fact[] supporting the claim" whose imputed discovery triggers the limitation period is not the order vacating the state conviction, but the fact that forms the basis for the motion to vacate the state conviction. Thus, the one-year period begins to run on the date on which the prisoner, in the exercise of due diligence, could have discovered the facts supporting his challenge to the state conviction. Even if the triggering event were the imputed discovery of the court order, the date the order "could have been discovered through the exercise of due diligence" would not be the date the order was in fact obtained, but the date it could have been obtained. Under either view, petitioner's Section 2255 motion was untimely.

A. The "Fact[] Supporting The Claim" Whose Imputed Discovery Triggers The Limitation Period Under 28 U.S.C. 2255 Para. 6(4) Is The Fact That Forms The Basis For The Motion To Vacate The State Conviction, Not The Order Vacating The State Conviction

1. Petitioner's interpretation is inconsistent with the text of 28 U.S.C. 2255 para. 6(4)

a. Much of petitioner's brief is devoted to the argument that the court order vacating his state conviction is the "fact" that "support[s] the claim" in his Section 2255 motion. See Br. 18-29. Petitioner reasons that, if a conviction is a "fact," then vacatur of the conviction must be as well, and that, because Daniels v. United States, 532 U.S. 374 (2001), makes clear that a prisoner may not challenge his federal sentence under Section 2255 until the state conviction has been vacated, it is the vacatur of the conviction that "support[s]" the claim. Even if petitioner is correct that the vacatur is a fact supporting his claim, it does not follow that the vacatur of the conviction is the event from which the limitation period runs under paragraph 6(4). Under that provision, the limitation period commences when the fact supporting the claim "could have been discovered through the exercise of due diligence." That means that it is not the fact itself that triggers the limitation period, but the discovery-or, more precisely, the imputed discovery-of the fact. And a court order is not a fact that is "discovered" by the party who requests it.

As petitioner acknowledges (Br. 24), the noun "discovery" means "the act, process, or an instance of gaining knowledge of or ascertaining the existence of something previously unknown or unrecognized." Webster's Third New International Dictionary 647 (1993). Similarly, the verb "discover" means "to make known (something secret, hidden, unknown, or previously unnoticed)." Ibid. To "discover," therefore, is to find something that exists, but whose existence or location was previously unknown. "Discovery" is distinct from the process of generating or manufacturing something that previously did not exist. That trial counsel had a conflict of interest, that the prosecution withheld exculpatory evidence, or that (as petitioner claimed in his state-court motion) there was not a valid waiver of the right to counsel-each of these is a preexisting fact that a federal prisoner might discover. A court order vacating a conviction is not. It is something that is brought into existence as a result of the prisoner's own actions-the filing of a motion and request for relief.

"When interpreting a statute, [a court] must give words their 'ordinary or natural' meaning." Leocal v. Ashcroft, 125 S. Ct. 377, 382 (2004) (quoting Smith v. United States, 508 U.S. 223, 228 (1993)). As the First Circuit explained in Brackett v. United States, 270 F.3d 60 (2001), cert. denied, 535 U.S. 1003 (2002), "[i]t would be an odd usage," not an ordinary or natural one, to say that a court order "could have been discovered" by a prisoner, because it is "obtained at [his] behest." Id. at 68. The court below reached the same conclusion. See J.A. 28 ("the vacatur of prior state convictions is a court action obtained at the behest of a federal prisoner, not 'discovered' by him").

b. Petitioner does not dispute that a party who takes the steps necessary to obtain a court order is not ordinarily said to "discover" it. He does contend, however (Br. 24), that the party who requests the order can "discover" it after the order has been issued. "It is conceivable," petitioner says, "that a defendant might not 'discover' a vacatur (or other court ruling) when it issues-e.g., if the prisoner is incarcerated and is not provided notice of the ruling." Ibid. That example may reflect a conceivable meaning of "discover," but for at least two reasons, it does not reflect the type of discovery contemplated by paragraph 6(4).

First, it would make little sense for the triggering date for a limitation period to be the date the issuance of a court order could have been discovered, since a court order is a public document that is readily available-particularly to the parties, on whom the order is served. Indeed, paragraph 6 of Section 2255 itself makes clear that, when the one-year limitation period runs from the date of a judicial decision, Congress intended the triggering date to be the date of the decision, not the date on which it could have been discovered by the prisoner. Paragraph 6(1) provides that the limitation period runs from "the date on which the judgment of conviction becomes final," and that date, if a certiorari petition is filed, is the date on which this Court "affirms a conviction on the merits on direct review or denies [the] petition for a writ of certiorari," Clay v. United States, 537 U.S. 522, 527 (2003). Paragraph 6(3) provides that the limitation period runs from "the date on which the right asserted was initially recognized by the Supreme Court, if that right has been newly recognized by the Supreme Court and made retroactively applicable to cases on collateral review." Paragraph 6 does not provide that the triggering date is the date on which the finality of the judgment of conviction, or this Court's initial recognition of the right asserted, could have been discovered through the exercise of due diligence. Accordingly, if the imputed discovery of the court order vacating a conviction triggered the limitation period under paragraph 6(4), as petitioner contends, court orders would be treated differently under paragraph 6(4) than under paragraphs 6(1) and 6(3). There is no reason to suppose that Congress intended such a result.5

Second, the triggering date under paragraph 6(4) is not the date on which the fact was discovered, but the date on which it "could have been" discovered "through the exercise of due diligence." The requirement that a prisoner exercise "due diligence" in discovering the factual predicate of his claim is one of the textual manifestations of Congress's intent that Section 2255 motions be prepared and filed expeditiously (a legislative purpose that is discussed in more detail below, see pp. 20-24, infra). If all the phrase "due diligence" meant in paragraph 6(4) was that, in a case of this type, a prisoner must exercise due diligence in discovering whether his motion has been decided, it would defeat that intent. Under petitioner's interpretation, a prisoner who took years to discover a factual basis for challenging a state conviction that could have been discovered within months, and then took years to file a motion to vacate the conviction that could have been filed within months, would be able to comply with the statute of limitations as long as he exercised "due diligence" in noticing the order that vacated his conviction and then filed his Section 2255 motion within a year. By requiring due diligence at the back end of the process (and only as to a "fact" of public record), but not at the front end (with respect to ordinary historical facts that can be discovered), petitioner's interpretation deprives the requirement of "due diligence" of virtually all its force. To avoid that result, the requirement should not be read to apply to the discovery of the court order.

2. Petitioner's interpretation undermines the purpose of the limitation provision of 28 U.S.C. 2255

a. As this Court has observed, and as petitioner acknowledges (Br. 33), "[t]here is no doubt" that Congress's purpose in enacting the AEDPA was "to further the principles of comity, finality, and federalism." Williams v. Taylor, 529 U.S. 420, 436 (2000). Accord Woodford v. Garceau, 538 U.S. 202, 206 (2003); Carey v. Saffold, 536 U.S. 214, 222 (2002). In Duncan v. Walker, 533 U.S. 167 (2001), the Court explicitly recognized that that purpose motivated the adoption of the AEDPA's one-year limitation period for habeas corpus petitions filed by state prisoners, 28 U.S.C. 2244(d)(1). The statute of limitations, the Court said, "quite plainly serves the well-recognized interest in the finality of state court judgments." 533 U.S. at 179. In particular, it "reduces the potential for delay on the road to finality by restricting the time that a prospective federal habeas petitioner has in which to seek federal habeas review." Ibid. While Duncan v. Walker involved the limitation provision for state prisoners, it is clear that the limitation provision for federal prisoners was motivated by the same legislative purpose, because the text of the state provision is nearly identical to that of the federal provision, see Clay v. United States, 537 U.S. at 528, and "the Federal Government, no less than the States, has an interest in the finality of its criminal judgments," United States v. Frady, 456 U.S. 152, 166 (1982).

As both the First Circuit in Brackett and the court below concluded, see 270 F.3d at 69-70; J.A. 29-30, petitioner's interpretation of paragraph 6(4) is fundamentally at odds with that purpose, because, far from "reduc[ing] the potential for delay on the road to finality," Duncan v. Walker, 533 U.S. at 179, petitioner's interpretation increases it. Under his view of paragraph 6(4), a prisoner can wait as long as he wants to seek vacatur of a prior state conviction, so long as the Section 2255 motion is filed within a year of the date on which the state conviction is vacated. That approach effectively eliminates any meaningful statute of limitations for motions of this type.

Moreover, petitioner's reading actually creates incentives to delay. "As time passes, and certainly once a state sentence has been served to completion, the likelihood that trial records will be retained by the local courts and will be accessible for review diminishes substantially." Lackawanna County Dist. Attorney v. Coss, 532 U.S. 394, 403 (2001). The passage of time also increases the likelihood that the State will "show[] no interest" in defending a conviction for which the defendant has long since completed his sentence. United States v. Walker, 198 F.3d 811, 814 (11th Cir. 1999) (Hill, J., concurring). See also Brackett, 270 F.3d at 63 (state "agreed to the motions" to vacate seven- and nine-year-old convictions). Under petitioner's interpretation of paragraph 6(4), federal prisoners will thus have "incentives to delay * * * their challenges to their state court convictions, and particularly to wait until the state ha[s] destroyed the trial or plea records, thus making it easier in some instances to obtain an order vacating the conviction." Brackett, 270 F.3d at 70. And the longer the delay in challenging a state conviction, the longer the delay in filing the Section 2255 motion. Indeed, by creating an incentive to delay challenges to state convictions, petitioner's interpretation "work[s] against [the] finality" of state-court judgments as well as federal ones. Ibid.6

b. Petitioner does not claim that his interpretation of Section 2255's limitation provision furthers "the well-recognized interest in the finality of [criminal] judgments." Duncan v. Walker, 533 U.S. at 179. Instead, petitioner contends that the court of appeals "plac[ed] undue emphasis on finality to the exclusion of comity and federalism" (Br. 34), and that his interpretation of paragraph 6(4) serves the principles of comity and federalism. Quoting Judge Barkett's dissent from the denial of rehearing en banc, petitioner argues as follows:

The Supreme Court has stated that in the habeas context a federal court should not "deprive [a] state-court judgment of its normal force and effect." Daniels, 532 U.S. at 378. This is exactly what the panel-majority does by stripping the state court of the ability to exculpate criminal defendants.

Br. 34-35 (quoting J.A. 52-53).

That view is misguided. Daniels said that a state-court judgment would be deprived of its "normal force and effect" if the Court countenanced "collateral attacks" on prior state convictions in Section 2255 motions in federal court. 532 U.S. at 378 (quoting Custis v. United States, 511 U.S. 485, 497 (1994)). Unlike a successful collateral attack on a conviction, rejection of petitioner's interpretation of paragraph 6(4) will not invalidate the judgment that vacated his conviction, and indeed will have no effect on that judgment at all. The court of appeals' decision thus does not prevent state courts from "exculpat[ing] criminal defendants." Pet. Br. 35 (quoting J.A. 53).

At bottom, petitioner seems to be arguing that it would violate principles of comity and federalism for a federal court to refuse a request to reduce a federal sentence after a state court has vacated a state conviction used to enhance the federal sentence. See Br. 34-35, 37. But the effect of the vacatur of a state-court conviction on a federal sentence, and any limitations on a federal prisoner's ability to seek a reduction of the sentence, are matters of federal, not state, concern. That principle is manifest in the fact that federal sentences may rely on vacated state convictions unless the basis for the vacatur fits within certain specified categories. See, e.g., 18 U.S.C. 3559(c)(7) (defendant sentenced under federal "three strikes" statute is entitled to resentencing only if prior conviction that was basis for sentence "is found * * * to be unconstitutional or is vitiated on the explicit basis of innocence"); Sentencing Guidelines § 4A1.2, comment. (n.10) (no prohibition on enhancing federal sentence on basis of prior conviction vacated by state court if conviction was set aside "for reasons unrelated to innocence or errors of law"); United States v. Cox, 83 F.3d 336, 339-340 (10th Cir. 1996) (reopening of federal sentence turns on district court's determination of basis for state court's vacatur of prior convictions).

Indeed, the federal interest in the vacatur of a long-since-served state conviction in many cases is much greater than any remaining state interest. The federal interest is what motivates the prisoner to file, and the lack of any remaining state interest may lead state officials not to contest, a motion to vacate a prior state conviction. Under those circumstances, it makes no sense not to have an effective federal statute of limitations.7

3. Under the correct interpretation of paragraph 6(4), the limitation period runs from the date on which the factual basis for the motion to vacate the state conviction could have been discovered with due diligence

If the vacatur of a state conviction is not the fact whose imputed discovery triggers the limitation period under paragraph 6(4), there remains the question of when the limitation period does begin to run. In rejecting the claim that petitioner makes here -that "the operative date under [paragraph 6(4)] is * * * the date the state conviction was vacated"-the First Circuit in Brackett held that the operative date is "the date on which the defendant learned, or with due diligence should have learned, the facts supporting his claim to vacate the state conviction." 270 F.3d at 68. That holding is correct. Unlike the order granting the motion to vacate the state conviction, the fact that forms the basis for the motion (in this case, an invalid waiver of the right to counsel) is both a fact that is "discovered" and one that, if pursued with "due diligence," will ordinarily lead to the prompt filing of a Section 2255 motion.

Petitioner does not dispute that the factual basis for the state-court claim is a "fact" susceptible of being "discovered," and he does not dispute that, in the context of Section 2255's limitation provision, the concept of "due diligence" makes more sense when applied to discovery of that fact than when applied to discovery of a court order. He does contend, however (Br. 21, 28-29), that, under this Court's decision in Daniels, the facts that form the basis for the state-court motion do not by themselves entitle a prisoner to relief under Section 2255, and that they therefore cannot be the facts that "support[] the claim" in the Section 2255 motion. For that reason, petitioner says, the facts that form the basis for the state-court motion are not the facts whose imputed discovery triggers the limitation period under paragraph 6(4).

Although a fact that forms the basis for challenging the state conviction is not a sufficient condition for challenging the federal sentence, it is a necessary condition, and in that sense it does support the claim. Indeed, because a state-court vacatur does not automatically justify federal resentencing, the order alone is not the sole "fact" that supports the motion under Section 2255-it is necessary to consider the underlying factual basis for the state-court order. See pp. 23-24, supra. At a minimum, the view that the factual basis for the state-court motion is the "fact[] supporting the claim" (even if only in a necessary, but not sufficient, way) is more compatible with the statutory text than the view that the court order requested by the prisoner is a fact "discovered" by him or that the prisoner must exercise "due diligence" only in learning whether his state-court motion has been decided. And even if the two views are equally compatible with the text, petitioner's interpretation should be rejected because it is fundamentally at odds with the legislative purpose. Cf. Duncan v. Walker, 533 U.S. at 178 (considering "the competing constructions in light of [the] AEDPA's purposes" and adopting the construction that "respect[s] the interest in the finality of state court judgments" rather than the one that "hold[s] greater potential to hinder finality").8

4. The government's interpretation does not prevent a prisoner who acts diligently from challenging a federal sentence after a state conviction has been vacated

Petitioner contends that, unless his interpretation of paragraph 6(4) is adopted, it will be "virtually impossible" (Br. 32) for a prisoner in his position to comply with the statute of limitations, because of the delay between the filing of a motion to vacate the prior state conviction and the issuance of an order granting the motion. Petitioner argues (Br. 30-33) that it is unfair to penalize a prisoner for events that are beyond his control. He also argues (Br. 38; accord NACDL Br. 12-13) that the consequence of rejecting his interpretation is that district courts will be clogged with "placeholder" Section 2255 motions that are filed before the state-court motion is decided, to guard against the possibility that the motion will be decided after the limitation period has expired. These concerns are misplaced. If the Court agrees with the government's interpretation of paragraph 6(4), there is no reason to think that prisoners who exercise due diligence in taking the actions over which they have control-discovering the factual basis for the state-court motion, filing the state-court motion, and filing the federal-court motion-will be unable to file a Section 2255 motion within the one-year limitation period. There is likewise no reason to think that "placeholder" motions will be required.

a. The factual basis for a challenge to a state-court conviction is ordinarily known (or knowable) during the guilty plea or trial, at the time the asserted violation occurs. As Judge Easterbrook has explained, the AEDPA's one-year limitation period "begins when the prisoner knows (or through diligence could discover) the important facts, not when the prisoner recognizes their legal significance." Owens v. Boyd, 235 F.3d 356, 359 (7th Cir. 2001); accord Hasan v. Galaza, 254 F.3d 1150, 1154 n.3 (9th Cir. 2001). Cf. United States v. Kubrick, 444 U.S. 111 (1979) (same rule for Federal Tort Claims Act's statute of limitations). For that reason, a defendant whose claim (for example) is that his guilty plea was not knowing and voluntary because the trial court failed to advise him of a particular right, or that his attorney rendered ineffective assistance by pursuing a particular trial strategy, will usually know the factual basis for the claim immediately.9 In most cases of that type, the facts that form the basis for the claim will thus be discoverable years before the federal conviction becomes final-indeed, years before federal charges are even brought. A deadline of a year from the date the federal conviction becomes final, therefore, will almost always provide ample time to seek and obtain an order vacating the state conviction.10

Even when the factual basis for a challenge to a state conviction is not discoverable at the time of the violation, it may still be discoverable at a sufficiently early date that the defendant will have an adequate opportunity to obtain the vacatur of the conviction within a year of the date his federal conviction becomes final. In Custis v. United States, 511 U.S. 485 (1994), for example, the factual basis for the state-court motion was known no later than the date of the federal sentencing. See id. at 488. In a case of that type, the prisoner will have at least the length of the federal direct-appeal process (which often lasts more than a year), plus an additional year, within which to obtain the vacatur of his state conviction.

b. In a case in which the factual basis for the challenge to the state conviction is discoverable only after the federal conviction becomes final, so that the limitation period is triggered by the imputed discovery, the prisoner may be able to obtain vacatur of the state conviction, and then file a Section 2255 motion, within a year of the date on which the basis for the claim could have been discovered, because state courts often take only a few months to decide a post-conviction motion. In Brackett, for example, the motion was filed in January 2000, see 270 F.3d at 63, and decided in June 2000, see Gonzalez v. United States, 135 F. Supp. 2d 112, 117 (D. Mass.), aff'd, 270 F.3d 60 (1st Cir. 2001), cert. denied, 535 U.S. 1003 (2002).11 In such a case, the Section 2255 motion will be timely under paragraph 6(4).

c. In a case in which the state court's delay in deciding the motion prevents a prisoner from obtaining an order vacating his state conviction within a year of the date the federal conviction became final or the date the factual basis for the state claim was discoverable, the prisoner may be eligible for equitable tolling. As the court below explained, that doctrine may be available when a party's filing is untimely because of extraordinary circumstances "beyond his control." J.A. 34. Delay by the state court may satisfy that standard in a case of this type if the prisoner acted diligently in taking the actions that are within his control. This Court has not considered "the availability of equitable tolling" of the AEDPA's limitation period, Duncan v. Walker, 533 U.S. at 181, but two members of the Court have expressed the view that there is "[no]thing in the text or legislative history of [the] AEDPA" that precludes it, id. at 183 (Stevens, J., joined by Souter, J., concurring in part and concurring in the judgment). And every court of appeals to consider the question has concluded that the AEDPA's limitation period is subject to equitable tolling in an appropriate case. See, e.g., Neverson v. Farquharson, 366 F.3d 32, 41 (1st Cir. 2004) (citing cases).

Petitioner contends that the use of equitable tolling "implicitly assumes that Congress wrote a flawed statute," and that equitable tolling "is necessary to fix it." Br. 33. That contention is mistaken, both because the availability of equitable tolling does not suggest that the basic statute of limitations is flawed and because equitable tolling will rarely be necessary if the government's interpretation of paragraph 6(4) is adopted. Cases involving the vacatur of a prior conviction are a small subset of the cases covered by that provision (because the newly discovered evidence is usually the direct basis for the Section 2255 motion), and cases in which equitable tolling may be appropriate are a small subset of that subset (because the factual basis for the state-court claim is usually discoverable many years before the federal conviction becomes final).12

B. Even If The "Fact[] Supporting The Claim" Whose Imputed Discovery Triggers The Limitation Period Under 28 U.S.C. 2255 Para. 6(4) Is The Order Vacating The State Conviction, The Date The Order "Could Have Been Discovered Through The Exercise Of Due Diligence" Is The Date The Order Could Have Been Obtained

1. Under petitioner's interpretation of paragraph 6(4), the relevant "fact[] supporting the claim" in a case of this type is the state court's order, and the triggering date for the one-year limitation period is the date the order "could have been discovered through the exercise of due diligence" after the order in fact came into existence. See Br. 24. (That will almost always be the date the order did come into existence.) But there is no reason why the limitation period should begin to run only after the relevant fact has come into existence. In a case in which the prisoner was able to control when the fact came into existence, it is possible that, if the prisoner had exercised due diligence, the fact could have come into existence, and thus have been discovered, at an earlier date. If petitioner is correct that the triggering fact is the court order vacating the state conviction, therefore, the limitation period should run, not from the date the court order was obtained, but from the date it could have been obtained if the prisoner had been diligent. For example, if a prisoner discovered the necessary facts a year later than he could have discovered them, and then waited an additional two years to file a motion to vacate the state conviction that could have been filed within a year of the discovery, the order granting the motion could have been obtained, and thus discovered, two years earlier if the prisoner had exercised due diligence.

This alternative interpretation is consistent with the text of paragraph 6(4), and it furthers the purpose of Section 2255's limitation provision by requiring prisoners to exercise diligence in obtaining the state-court order that is necessary before a Section 2255 motion can be filed. It also overcomes petitioner's two principal objections to the government's other interpretation: it does not presume that the "facts supporting the claim" in the Section 2255 motion are the facts that form the basis for the state-court motion; and it cannot be criticized on the ground that the limitation period may expire while the state-court motion is pending, because the period during which the motion is pending is excluded from the calculation.13

2. There are three possible objections to this alternative interpretation of paragraph 6(4). If the Court rejects the government's primary contention, none of these objections justifies rejecting the alternative suggested here.

The first possible objection is that the court order that could have been obtained if the state claim had been pursued diligently is not the "fact[] supporting the claim," because it would have been a different order than the one that was obtained. An earlier order, for example, would have had a different date and docket number, and might have used different language, than the later one. But if petitioner is correct that the "fact[] supporting the claim" in a case of this type is "the vacatur of [his] state conviction" (Br. 29), then the relevant fact whose imputed discovery triggers the limitation period is just that: a judicial declaration of invalidity. It is not the particular document in which the declaration happens to be made.

The second possible objection is that a prisoner who acted diligently in pursuing his state claim might not have succeeded in having the conviction vacated, because, for example, the records might not yet have been destroyed or the State might still have had an interest in defending the conviction. But as long as a legal basis for relief was available at the time the prisoner could have filed the motion in the exercise of due diligence, it should be presumed that he would have obtained the relief if he had requested it. A prisoner should not be heard to complain that he obtained relief at a later date for reasons unrelated to the merits of the claim, because, just as "[a] defendant has no entitlement to the luck of a lawless decisionmaker," Lockhart v. Fretwell, 506 U.S. 364, 370 (1993) (quoting Nix v. Whiteside, 475 U.S. 157, 175 (1986)), he has no entitlement to the luck of missing records or of a prosecutor uninterested in defending the conviction.

The third possible objection is that this interpretation will require federal courts to determine not only when the factual basis for the state claim could have been discovered, but also when the state-court motion could have been filed and when the motion would have been decided if it had been filed earlier, and that the second and third determinations are too speculative. This objection is also without merit. As to the question of when the state-court motion could have been filed, prisoners can be given a full year from the date on which the factual basis for the motion could have been discovered through the exercise of due diligence, because paragraph 6(4) reflects a congressional judgment that, for the ordinary type of collateral attack on a conviction based on newly discovered evidence, it should always be possible to file the motion within a year of the date on which the evidence could have been discovered. As to the question of when the motion would have been decided, except perhaps in unusual cases, federal courts can presume that, if the motion had been filed earlier, the state court would have taken the same amount of time to decide the motion that it took when the motion was in fact filed.

C. Under Either Of The Permissible Interpretations Of 28 U.S.C. 2255 Para. 6(4), Petitioner's Section 2255 Motion Was Untimely

1. If the "facts supporting the claim" whose imputed discovery triggers the limitation period under paragraph 6(4) are the facts that form the basis for the motion to vacate the state conviction, petitioner's Section 2255 motion was untimely. In this case, the triggering date for the one-year limitation period under paragraph 6(1) was April 24, 1996, the AEDPA's effective date, because petitioner's judgment of conviction became final before that date. See note 2, supra. The triggering date under paragraph 6(4) was earlier, because petitioner was aware of the factual basis for his state claim at the time of his direct appeal (at the latest), which predated the AEDPA's effective date. See J.A. 7-8, 33. Petitioner was therefore required to file his Section 2255 motion by April 24, 1997. He did not do so until February 13, 2001, nearly four years too late.

Nor could petitioner's motion be rendered timely by equitable tolling of the period during which his state-court motion was under consideration (February 6, 1998 through October 24, 2000). The one-year limitation period expired before the state-court motion was filed, and a motion that is "filed following the expiration of the limitations period cannot toll that period," because "there is no period remaining to be tolled." Webster v. Moore, 199 F.3d 1256, 1259 (11th Cir.) (per curiam), cert. denied, 531 U.S. 991 (2000). As a consequence, even if the 32 months during which the state motion was pending were subtracted from the 58 months between the date the limitation period commenced and the date the Section 2255 motion was filed, the difference would be 26 months, and petitioner's federal motion would still be 14 months late.

2. Petitioner's Section 2255 motion was also untimely if the "fact[] supporting the claim" whose imputed discovery triggers the limitation period is the state-court order vacating the conviction, because the order in this case could have been obtained (and thus discovered) in the exercise of due diligence more than a year before the Section 2255 motion was filed. Since the factual basis for petitioner's state-court claim was that he did not validly waive his right to counsel when he pleaded guilty, he likely was (or should have been) aware of the basis for the claim at the time of his plea. See, e.g., Brackett, 270 F.3d at 71. Even giving him the benefit of the doubt, however, he certainly was aware of it by the time the court of appeals rejected the same claim in affirming his sentence in December 1995. See J.A. 7-8, 33. If petitioner had filed his state-court motion within a year of that date (i.e., by December 1996), the motion could have been decided by August 1999 (assuming the state court took the same amount of time it took when the motion was actually filed). The triggering date under paragraph 6(4) was thus no later than August 1999, which was the applicable date because the AEDPA's effective date (the triggering date under paragraph 6(1)) was earlier. Under the alternative interpretation of paragraph 6(4), therefore, petitioner was required to file his Section 2255 motion by August 2000. He did not do so until February 2001, at least six months too late.

CONCLUSION

The judgment of the court of appeals should be affirmed.

Respectfully submitted.

PAUL D. CLEMENT
Acting Solicitor General
CHRISTOPHER A. WRAY
Assistant Attorney General
MICHAEL R. DREEBEN
Deputy Solicitor General
DAN HIMMELFARB
Assistant to the Solicitor
General
NINA GOODMAN
Attorney

DECEMBER 2004

1 In both cases, the Court made an exception for convictions that were uncounseled in violation of Gideon v. Wainwright, 372 U.S. 335 (1963). See Daniels, 532 U.S. at 382; Custis, 511 U.S. at 487. The plurality in Daniels also left open the possibility that Section 2255 would be available in "rare cases in which no channel of review was actually available to a defendant with respect to a prior conviction, due to no fault of his own." 532 U.S. at 383 (opinion of O'Connor, J.).

2 The AEDPA took effect on April 24, 1996, but in cases (like this one) in which the judgment of conviction became final before that date, the courts of appeals have uniformly afforded the prisoner one year from the AEDPA's effective date within which to file a Section 2255 motion. See, e.g., Rogers v. United States, 180 F.3d 349, 353-354 & n.9 (1st Cir. 1999) (citing cases), cert. denied, 528 U.S. 1126 (2000).

3 The AEDPA also added a statute of limitations, codified at 28 U.S.C. 2244(d)(1), for habeas corpus petitions filed by state prisoners. It is identical in substance, though not in its language, to the statute of limitations for federal prisoners. The state prisoners' counterpart of the provision at issue here is 28 U.S.C. 2244(d)(1)(D), which states that the one-year limitation period runs from "the date on which the factual predicate of the claim or claims presented could have been discovered through the exercise of due diligence."

4 In an opinion dissenting from the denial of rehearing en banc (J.A. 50-53), Judge Barkett agreed with the views expressed by Judge Roney in his panel dissent and by the Fourth Circuit in Gadsen.

5 Paragraph 6 as a whole is thus consistent with the understanding that judicial orders and decisions are not the kind of "facts" that can be discovered for purposes of Section 2255's statute of limitations. Rather, litigants are presumed to know about orders and decisions, and federal courts need not entertain litigation about whether the "fact" of a helpful precedent could have been discovered through the exercise of due diligence.

6 Petitioner's interpretation not only works against the purpose of Section 2255's limitation provision, it works against the purpose of federal laws like the ACCA and the Guidelines' career-offender provision, which is to ensure that "career" criminals receive the most severe punishment. See, e.g., 28 U.S.C. 994(h) (directing Sentencing Commission to ensure sentences for career offenders "at or near the maximum term authorized"). When a prisoner who delays a challenge to a state conviction is successful in having the conviction vacated, his success is often attributable to the fact that the State is no longer able or willing to defend the conviction, not to the fact that the prisoner is innocent or that his conviction was unlawfully obtained. If the prisoner is then able to use the vacatur of his state conviction to obtain a reduced federal sentence, the result will be a lower sentence than Congress intended for the worst recidivists.

7 In light of that more immediate federal interest, it would be particularly anomalous to allow a defendant to take an unlimited amount of time before commencing a challenge to his state conviction, with no federal time limit. A federal conviction may also be used to support a recidivist sentence enhancement, and any collateral challenge to the validity of a prior federal conviction would have to meet the time limits set forth in the AEDPA. Thus, petitioner's view leads to the conclusion that, despite the government's paramount interest in achieving finality of a federal recidivist sentence and the AEDPA's regulation of the time period for challenging a prior federal conviction, there would be no federal time period within which a defendant would have to commence a challenge to a prior state conviction. There is no reason to read the AEDPA to give the defendant unilateral control, in that one instance alone, over when to undertake a legal challenge that could lead to upsetting the finality of the federal judgment.

8 If the Court agreed with petitioner that the factual basis for the state-court motion is not the "fact[] supporting the claim" in the Section 2255 motion, it would not necessarily follow that petitioner's interpretation of paragraph 6(4) is correct. One might conclude that paragraph 6(4) simply does not apply in cases of this type, and that such cases are always governed by paragraph 6(1) (unless paragraph 6(2) or 6(3) supplies a later triggering date). The court below rejected petitioner's contention that the limitation period under paragraph 6(4) runs from the date of the state court's decision, but it is not clear whether the court believed that the limitation period runs from the imputed discovery of the factual basis for the state-court motion or that paragraph 6(4) is inapplicable in cases of this type. There was no need for the court to decide that question, because, even if it believed that paragraph 6(4) applies, petitioner was aware of the factual basis for his state-court motion before his federal conviction became final, see J.A. 33, and the "latest" date under paragraph 6 of Section 2255 was thus the date on which the conviction became final.

9 See, e.g., Brackett, 270 F.3d at 71 (if the prisoner "received an inadequate colloquy" when he pleaded guilty, "he knew it then"); Owens, 235 F.3d at 359 ("the principal fact setting the stage for the current ineffective-assistance claim-that [the prisoner's] trial counsel attempted to present a [meritless] coercion defense-was known at trial"); Candelaria v. United States, 247 F. Supp. 2d 125, 130 (D.R.I. 2003) ("the 'facts' which matter * * * are those which existed at the time of the plea colloquy-namely, that [the prisoner] was not advised of the factual basis of the charges against him, that he was not informed of the state's burden of proof, that he was not told which constitutional rights he would forego by pleading guilty, and that he was not notified that he was facing deportation upon entering a guilty plea").

10 See, e.g., Brackett, 270 F.3d at 62-63 (factual bases for challenging two state convictions were known in 1991 and 1993; federal conviction became final in 1998); Candelaria, 247 F. Supp. 2d at 126-130 (factual basis for challenging state conviction was known in 1992; federal conviction became final in 2001).

11 See also United States v. LaValle, 175 F.3d 1106, 1107 (9th Cir. 1999) (motion was filed in September 1996 and granted in January 1997); Candelaria, 247 F. Supp. 2d at 126-127 (motion was filed in June 2001 and granted in November 2001).

12 As the court below observed (J.A. 31 n.5, 37), the result in United States v. Gadsen, 332 F.3d 224 (4th Cir. 2003), which endorsed the interpretation of paragraph 6(4) advanced by petitioner, might well have been the same if the Fourth Circuit had rejected that interpretation but applied equitable tolling. In Gadsen, the motion to vacate the state conviction was filed no more than ten months after the factual basis for the motion could have been discovered and more than 18 months before the prisoner's federal conviction became final; the state conviction was vacated nearly 18 months after the federal conviction became final; and the Section 2255 motion was filed within a year of the state court's decision. See id. at 225-226. Under those circumstances, it might have been appropriate to toll the period from the date the federal conviction became final (the triggering date for the limitation period) through the date the state courts acted.

13 In this connection, it bears noting that the Section 2255 motion filed by the prisoner in Gadsen was likely timely under this interpretation as well. The state-court motion in that case was filed ten months after the prisoner's no-contest plea in state court; the conviction was vacated 36 months after the motion was filed (and nearly 18 months after the federal conviction became final); and the Section 2255 motion was filed 11 months after the conviction was vacated. See 332 F.3d at 225-226. If the Fourth Circuit had adopted the alternative interpretation of paragraph 6(4) proposed here, it might well have concluded that Gadsen acted diligently in filing his state-court motion; that the state-court order thus could not have been obtained any earlier than the date it was issued; and that the Section 2255 motion was therefore filed within a year of the date the order was discoverable in the exercise of due diligence.